Response generation systems are a corpus-based chatbot's methods of creating responses via retrieval or generation methods.

Response Generation Systems

Corpus-based chatbots are chatbots that draw their conversational data from corpuses, or vast linguistic datasets, of human conversation from various sources like social media and TV shows. 

University of California, Santa Cruz

Google

Chatbots are designed for unstructured conversation, and are meant to mimic natural back and forth human dialogue

Chatbots

An on-going but a helpful book resource about NLP
https://web.stanford.edu/~jurafsky/slp3/

Speech and Language Processing (3rd ed. draft) 

Rule-based chatbots are chatbots that create their conversations from set rules, such as responding "What makes you think I _____ you" in response to the statement "You _____ me." Early notable rule-based chatbots include ELIZA, which essentially acted as a therapist to clinical psychology patients, and PARRY, which modeled schizophrenic behavior using responses based on variables for fear and anger.

Rule-Based Chatbots

Corpus-Based Chatbots

Chatbots are typically evaluated by humans, who assign the chatbot a score that weighs different characteristics of performance such as engagingness,  fluency, humanness, etc. Chatbots can be evaluated by either a user (participant evaluation) or a third-party observer (observer evaluation).

Evaluating Chatbots

Chatbot's architecture can be hybrid ,taking advantage of both rule-based and neural/corpus based systems sometimes they even utilize elements from frame-based structure.

Hybrid Chatbots

ELIZA is an early natural language processing system that could carry on a limited conversation with a user by imitating the responses of a Rogerian psychotherapist.

Learn Before

Related

Learn After